tensor<20x20xf32>, none,
        tensor<1x20xf32>, tensor<1x20xf32>,
        none, none, none, none) -> tensor<1x28x20xf32>
    %1 = "quantfork.stats"(%0) {layerStats = dense<[-1.0, 2.0]> : tensor<2xf32>} : (tensor<1x28x20xf32>) -> tensor<1x28x20xf32>
    func.return %1 : tensor<1x28x20xf32>

        none, none,
        tensor<1x20xf32>, tensor<1x20xf32>,
        none, none, none, none) -> tensor<1x28x20xf32>
    %1 = "quantfork.stats"(%0) {layerStats = dense<[-1.0, 2.0]> : tensor<2xf32>} : (tensor<1x28x20xf32>) -> tensor<1x28x20xf32>
    func.return %1 : tensor<1x28x20xf32>
// CHECK-DAG: %[[none:.*]] = "tfl.no_value"() <{value}> : () -> none

  // CHECK-LABEL: func @parallel_execute_with_tiled_input
  // CHECK-SAME: (%[[ARG_0:[a-z0-9]*]]: tensor<128x10xf32>, %[[ARG_1:[a-z0-9]*]]: tensor<128x10xf32>, %[[ARG_2:[a-z0-9]*]]: tensor<*xi32>, %[[ARG_3:[a-z0-9]*]]: tensor<*xi32>)
  func.func @parallel_execute_with_tiled_input(%arg0: tensor<128x10xf32>, %arg1: tensor<128x10xf32>, %arg2: tensor<*xi32>, %arg3: tensor<*xi32>) -> (tensor<*xi32>, tensor<*xi1>) {
    // CHECK: tf_device.replicate

// CHECK-LABEL: @ShardingAttr
func.func @ShardingAttr(%arg0: tensor<128x10xf32> {mhlo.sharding = "\08\03\1A\02\01\02\22\02\00\01"}, %arg1: tensor<10x1024xf32> {mhlo.sharding = "\08\01\1A\01\01\22\01\00"}, %arg2: tensor<128x1024xf32> {mhlo.sharding = ""}) -> (tensor<128x10xf32>, tensor<10x1024xf32>, tensor<128x1024xf32>) {

  func.return %2, %3 : tensor<128x16xf32>, tensor<128xi32>
}

// CHECK-LABEL: PruneUnusedLstm
func.func @PruneUnusedLstm(%arg0: tensor<1x28x28xf32>) -> (tensor<1x28x28xf32>) {
    %input = "tfl.quantize"(%arg0) {qtype = tensor<1x28x28x!quant.uniform<i8:f32, 0.003:-128>>} : (tensor<1x28x28xf32>) -> tensor<1x28x28x!quant.uniform<i8:f32, 0.003:-128>>

func.func @expandDims(%arg0: tensor<2x2xf32>, %arg1: tensor<i32>) -> tensor<1x2x2xf32> {
  %0 = "tf.ExpandDims"(%arg0, %arg1) : (tensor<2x2xf32>, tensor<i32>) -> tensor<1x2x2xf32>
  func.return %0 : tensor<1x2x2xf32>

// CHECK-LABEL:expandDims
// CHECK:  "tfl.expand_dims"(%arg0, %arg1) : (tensor<2x2xf32>, tensor<i32>) -> tensor<1x2x2xf32>
}

func.func @squeezeDefault(%arg0: tensor<1x2x2xf32>) -> tensor<2x2xf32> {

  %2 = "tf.BatchToSpaceND"(%arg0, %0, %1) {device = ""} : (tensor<3x5x2xf32>, tensor<1xi32>, tensor<1x2xi32>) -> tensor<1x8x2xf32>

  // CHECK: return [[VAL8]] : tensor<1x8x2xf32>
  func.return %2 : tensor<1x8x2xf32>
}

func.func @fake_quant_with_min_max_args(%arg0 : tensor<?x?xf32>) -> tensor<?x?xf32> {

func.func @div_dense_different_rank() -> tensor<1x2x2xf32> {
  %cst_0 = arith.constant dense<[[[1.0],[2.0]]]> : tensor<1x2x1xf32>
  %cst_1 = arith.constant dense<[[2.0, 3.0]]> : tensor<1x2xf32>

  %0 = "tfl.div"(%cst_0, %cst_1) {fused_activation_function = "NONE"} : (tensor<1x2x1xf32>, tensor<1x2xf32>) -> tensor<1x2x2xf32>

  func.return %0 : tensor<1x2x2xf32>

      %2 = "tf.Cast"(%arg0) {Truncate = false} : (tensor<1x8x2xi32>) -> tensor<1x8x2xf32>
      tf_device.return %2 : tensor<1x8x2xf32>
    // CHECK: () -> tensor<1x8x2xf32>
    }) {device = "/device:CPU:0"} : () -> tensor<*xf32>
    // CHECK: "tf.Cast"(%{{.*}}) <{Truncate = false}> : (tensor<1x8x2xf32>) -> tensor<*xf32>
    // CHECK: (tensor<i32>, tensor<1x8x2xf32>) -> (tensor<1x8x1xf32>, tensor<1x8x1xf32>)